ClockTutorial #2a: Forty-Five Years of Pittendrigh's Empirical Generalizations

i-710d005c8660d36282911838843a792d-ClockWeb logo2.JPG

This is the third in the series of posts designed to provide the basics of the field of Chronobiology. This post is interesting due to its analysis of history and sociology of the discipline, as well as a look at the changing nature of science. You can check out the rest of Clock Tutorials here.

It appears that every scientific discipline has its own defining moment, an event that is touted later as the moment of "birth" of the field. This can be a publication of a paper (think of Watson and Crick) or a book ("Origin of Species" anyone?). In the case of Chronobiology, it was the 1960 meeting at Cold Spring Harbor. The book of Proceedings from the Meeting (Symposia on Quantitative Biology, Vol.XXV) is a founding document of the field: I have two copies, my advisor has three (all heavily used and annotated).

The 1960 meeting was not the first one. There were a few others before, e.g., one in Stockholm, Sweden, another in Feldafing, Germany. But the Cold Spring Harbor meeting was special. Why? I don't know - I wasn't even born yet. I have a hunch that there were several aspects of this symposium that made it different from the preceding meetings. First, the sheer number of participants was larger thus, perhaps, reaching a critical mass, or crossing a treshold needed for the group to feel as if they are not just congregating individuals but a part of something bigger. Additionally, being a part of a venerable tradition of powerful meetings at Cold Spring Harbour may have signalled to the group that they were finally taken seriously by a broader scientific community.

Second, it appears that this was the first time all the participants really understood that the diverse phenomena they were studying were unified by more than just appearance of oscillations, but that they were different aspects of just one basic biological phenomenon for which, for the first time, they had a name: circadian rhythms and clocks (as well as other circa- rhythms), the term coined by Franz Halberg in 1959.

It is fun to read the Proceedings every now and then and compare the state of science, as well as way of thinking, between 1960 and 2005. One obvious and expected difference is in genetics. Today almost every paper has a picture of a gel. In 1960. Watson and Crick were still busy. The lack of understanding of molecular (the often-used term then was "subcellular") mechanism was a frequent lament in the papers comprising the Proceedings, and was subsituted with an unusual (to our eyes) amount of complex mathematical modelling (and that was before the personal computers!).

Another thing that immediatelly catches one's attention is the enormous number of species studied compared to current reliance on about a dozen model organisms (e.g., human, rat, mouse, hamster, chicken pineal in vitro, Xenopus frog, zebrafish, fruitfly, Neurospora bread mold, Arabidopsis plant, Synechococcus cyanobacteria). Particularly, the number of plants and protists is amazing, as the current chronobiology is so animal-centered, mainly due to the need to use models for human disease for the purpose of funding.

If you ask researchers today who the pioneers of the field are, the most likely list of "Fathers of Chronobiology" will emerge, including Colin Pittendrigh, Jurgen Aschoff, Gustav Kramer, Erwin Bunning, Karl von Frisch, August Forell, Curt Richter, Frank Brown, Max Renner, Rutger Weaver, Woodie Hastings, Eberhard Gwinner, John Palmer, Franz Halberg, Michael Menaker and Serge Daan, among others (the last three are still active, the rest are retired or dead). If you asked the 1960 participants, including all of the mentioned men, they would have probably trotted out a list that looks like this: Ingeborg Beling, Anthonia Kleinhoonte, Rose Stoppel, Beatrice Sweeney, Patricia DeCoursey, Janet Harker, Miriam Bennett, Dorothea Minis and Ursula von StPaul. Huh? "Mothers of Chronobiology"? What happened?

The proportion of the women in chronobiology is today, as it apparently always was, very high, even for biology which is rating the best among natural sciences. I am not going to risk a "Summers" mistake and suggest anything remotely like a genetic explanation (e.g., cyclicity of woman's physiology attracting women to study cycles). Yet today, most of the Big Names are men, while in the past at least half of the Big Names were women. There are many women inhabiting the labs and doing marvelous research today, but rarely as heads of Big Labs. Patricia DeCoursey is still active and, a living legend as she is, she can do whatever she wants. Amita Seghal and Carla Green also come to mind as current big female stars with their own Big Labs. But otherwise it is men, men, men (hey, I am a man and I want to have a Big Lab and become a Big Name, too). Why has the situation changed over the decades?

Big Names today are people with Big Labs. Male-dominated culture results in more men heading Big Labs. At the same time, expense of research ensures that much of the work is neccessarily done in Big Labs. It is difficult nowadays to get exciting work funded, done, published and revered if your lab is a little dark room in the basement. Half a century ago, the picture was different. Science did not require exorbitant amounts of money, huge lab space, dozens of technicians and students, and rapid rate of publishing. One could spend years in a dark basement room and finally emerge with such exciting and novel discoveries as to immediatelly become a Big Name.

The male-centered culture ensured that best and brightest male students got into the biggest, most popular labs, leaving the female candidates with remnants - working with some semi-crazy professor in the basement who is doing some weird semi-scientific stuff. It is a big risk, but if that crazy professor is onto something revolutionary, the final payback can be huge. Until the 1960s, chronobiology was regarded as weird stuff. Gustav Kramer was partically funded by Duke University Department of Parapsychology! It was as far in the left field as you could get in science. It was the daring loners who did the best of the earliest stuff, not the thousands of mainstream scientists involved in doing regular science of the day in big laboratories.

Today, it is difficult being a weird loner. Only mainstream, bandwaggon science gets funded at all. Chronobiology is now as mainstream as can be. There is almost no issue of Science Magazine without at least one paper on the topic. Does it mean that it has lost some of its creativity? It is interesting to see, reading the Proceedings, what did the last 45 years bring to the field. While nothing was known about clock genes and the molecular mechanism, it was strongly anticipated and some fairly plausible models were put forward, not too different from what it turned out to be in the end.

On the other hand, it comes as a shock for the current researcher in the field to read the Proceedings and realize that practically everything we know now was already known in 1960. We added detail, lots of detail, but no new concepts, no new rules, no new creative experimental protocols. For instance, look at the list of Empirical Generalizations about biological clocks from the beginning of the paper by Colin Pittendrigh (probably the most cited paper in the field ever):

I. CRs (circadian rhtyhms) are defined as those rhythms whoe tauFR (freerunning period) is an approximation to the period of the earth's rotation. [This is a definition that is still used]

II. CRs are ubiquitous in living systems (both in terms of all organisms on Earth and in terms of all functions within an organism) [ We now know that many bacteria, as well as deep-oceanic organisms do not have clocks, while in some subterranean animals the clock has degraded somewhat. Every function in every body is under circadian control]

III. CRs are endogenous [Yes, they are generated within the organism, not by organisms responding to environmental cycles - this was a big controversy in the 1950s]

IV. CRs are self-sustaining oscillations [Yes, they still are]

V. CRs are innate [Yes. We have identified a whole suite of genes and understand how they work]

VI. CRs occur autonomously at all levels of organization [Yes, though we tend to concentrate on molecules today]

VII. The system displays remerkable precision [True]

VIII. Period is open to spontaneous or induced shifts within a range of values [Correct]

IX. Species differ in the range of relizable values of period [True]

X. Period may show aftereffects of the regime immediately preceding the steady-state freerun being studied [True, I will write more about this later]

XI. CRs are temprature-independent [True]

XII. Period is light-intensity dependent (Aschoff's Rule) [This is generally true, exceptions have been found, nobody is studying this phenomenon any more, I will write more about it later]

XIII. CRs are entrainable by a restricted class of environmental periodicities [True]

XIV. The phase of a freerunning CR can be shifted by single perturbations [True]

XV. Transients always precede attainment of a new steady-state [Exceptions have been found since then, but nobody is studying the phenomenon, I will write more about it later]

XVI. CRs have so far proved surprisingly intractable to chemical perturbation [True. Heavy water, lithium, melatonin and a couple of other hormones are the only substances that affect the clock].

As you can see, everything important about clocks has been figured out by 1960 and has stood the test of time since then.

[Colin S. Pittendrigh]

If anything, today's crop of students is less sophisticated - some of the papers from 1960 induce a headache, that is how conceptually difficult it is to understand them. Yet they are not bogus or useless. Those are very important papers in the field, resulting from years of very creative experimenting, and should be a part of training of every new chronobiologist. But, they are not taught in courses. They are not read at lab meetings. Much of the lore is now lost.

Those who forget history are liable to repeat it. Those who are unaware of the early research in the field can make big mistakes. I have seen a young researcher present his paper at a meeting and he tried to interpret his data with an explanation that was disproven about 50 years ago. Prompted by a person in the audience (who actually did that research 50 years ago), the young guy admitted he had never read or even heard of that series of papers. And this was not some obscure series of papers, this was doctoral research of one of the founding leaders of the field, something that should be everyone's required reading.

Reading the old literature (from late 1920s till late 1980s) can be an excerise in frustration. There are creative ideas and plausible hypotheses that nobody has tested yet. There are lines of research that have been abandoned, yet modern techniques should make the continuation easy and fun, but nobody is doing it. There are species that promise to be great models for modern research, but nobody is looking at them any more. Finally, there is a long list of papers with data that directly contradict the current understanding of the way the circadian clock works. Nobody is revisiting those data to repeat the experiments or explain them in light of current knowledge.

When one reads those papers, one starts to doubt if the current model is correct. The recent data are compatible with the present consensus model, but some of the old data are not. If all you know is the current model and the current data, and that is the only way you think about it, you will never realize that there may be alternative models that are compatible with ALL data, both recent and old. The recent research assumes that the model is correct and tries to work out the fine details. Knowledge of the old data should force people to directly test if the model is correct or not. But who is going to get funded to repeat a 1928 experiment?

I plan to devote a whole future post to the whole gammut of data contradiciting the current molecular model. But I first have to make my way to describing the model itself, and that will take several posts over the next few weeks.


More like this